Incorporation of calcium in glasses: a key to understand the vitrification of sewage sludge

HAL Id: hal-03208408

https://hal.univ-angers.fr/hal-03208408

Submitted on 26 Apr 2021

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Incorporation of calcium in glasses: a key to understand the vitrification of sewage sludge

Mariona Tarrago, Irene Royo, Maite Garcia- Valles, Salvador Martinez, Daniel Neuville

To cite this version:

Mariona Tarrago, Irene Royo, Maite Garcia- Valles, Salvador Martinez, Daniel Neuville. Incorporation of calcium in glasses: a key to understand the vitrification of sewage sludge. International Journal of Applied Glass Science, Wiley, 2021, �10.1111/ijag.15920�. �hal-03208408�

Elsevier Editorial System(tm) for Ceramics International

Manuscript Draft

Manuscript Number:

Title: Incorporation of calcium in basaltic glasses: a key to understand the vitrification of sewage sludge

Article Type: Full length article

Keywords: glass-ceramic, sewage sludge, calcium, valorization, viscosity, nucleation

Corresponding Author: Dr. Mariona Tarrago, Ph.D.

Corresponding Author's Institution: Institut de Physique du Globe de Paris

First Author: Mariona Tarrago, PhD

Order of Authors: Mariona Tarrago, PhD; Irene Royo, B.Sc.; Maite Garcia- Valles, Ph.D.; Salvador Martinez, Ph.D.; Daniel R. Neuville, Ph.D.

Abstract: The quantity of sewage sludge generated daily in water decontamination represents a major environmental problem. Remediation strategies focus on using the wastes as raw materials to reduce storage costs and minimize the need for mining. An approach by vitrification reduces the volume of waste and inertizes hazardous elements by binding them to the structure of chemically stable glasses and glass-ceramics.

The valorization process of sewage sludge by vitrification has been

simulated by producing a glass and a glass-ceramic from a basalt enriched in calcium that lies between the stability fields of pyroxene and

melilite in the system CaO-MgO-SiO2-Al2O3. Nucleation at the temperature of maximum nucleation rates (650 and 675 ºC) of this glass causes the formation of a biphasic system (crystal + glass) that constrains its rheological behavior enhancing the formation of a large amount of nuclei that result in a fine microstructure, forming a glass-ceramic. The

microhardness of the glass (8.2 GPa) and the glass-ceramic (8.6 GPa) and leaching tests (in the ppb range) place both the glass and the glass- ceramics at the high end of the mechanical properties and chemical resistance of ceramic tiles for the building industry.

Suggested Reviewers: Alexander Karamanov PhD

Professor, Department of Phase Formation, Crystalline and Amorphous Materials, Bulgarian Academy of Sciences

karama@ipc.bas.bg

Prof. Karamanov is a world-class expert in the vitrification of wastes and has extensively worked in the production process of glass-ceramics.

Ina Mitra PhD

Senior Scientist, Schott AG ina.mitra@schott.de

Dr. Ina Mitra is a Senior Scientist at the glass producer Schott AG (Germany), where she develops glasses and glass-ceramics for devices in

Pura Alfonso PhD

Professor, Mining Engineering and Natural Resources, Universitat Politècnica de Catalunya

pura@emrn.upc.edu

Prof. Alfonso has extensively worked in vitrification and glass-ceramics production as a means to give a new use to mining waste.

Opposed Reviewers:

Mariona TARRAGÓ Géomatériaux

Institut de physique du globe de Paris UMR 7154

+ 33 (0)1 83 95 7758 tarrago@ipgp.fr

Editor, Ceramics International Submission to Ceramics International

Paris 11th of March 2020

Dear Editor,

Please find enclosed our manuscript “Incorporation of calcium in basaltic glasses: a key to understand the vitrification of sewage sludge” by M. Tarragó, I. Royo, M. Garcia-Valles, S. Martínez and D.R. Neuville, which we would like to submit for publication as a peer-reviewed Article in the Ceramics International.

This manuscript brings new data on the inertization and valorization of sewage sludge simulated using a Ca-enriched basalt. The nucleation and crystal growth temperatures of all the phases have been carefully investigated in order to obtain a glass-ceramic with superior physicochemical properties than the parent glass. Our data on the stability of merwinite confirms its crystallization as a metastable phase.

We think our findings would appeal to the readership of the Ceramics International as vitrification is a means of reducing the storage issues and chemical hazards related to the landfilling of wastes. The results show potential of sewage sludge-like materials as inertization matrix for industrial wastes containing potentially toxic elements such as tannery (Cr) or galvanic sludge (Cu, Zn). The obtained glasses and glassceramics can be used as abrasion-resistant materials in construction.

We would like to suggest, if the Editor finds it appropriate, Prof. Alexander Karamanov, Dr. Ina Mitra, and Prof. Pura Alfonso as reviewers of our work. They all have extensive research experience in glass and crystallization processes in wastes.

This manuscript has not been published elsewhere nor is under consideration by another journal. All authors approve the manuscript and agree with its submission to the Ceramics International.

We are looking forward to receiving your decision about our work.

Yours sincerely, Mariona Tarragó Signature Cover Letter

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

Incorporation of calcium in basaltic glasses: a key to understand the vitrification of sewage sludge

M.Tarrago

*, I. Royo

, S. Martínez

, M. Garcia-Valles

and D.R. Neuville

1

Dpt., Mineralogia, Petrologia i Geologia Aplicada, Fac. de Ciències de la Terra. Universitat de Barcelona, c/ Martí i Franquès, s/n, 08028 Barcelona (Spain)

2

Géomatériaux, CNRS-Institut de Physique du Globe de Paris, Université de Paris, 1 rue Jussieu 75005 Paris (France)

E-mail tarrago@ipgp.fr

*Corresponding author Abstract

The quantity of sewage sludge generated daily in water decontamination represents a major environmental problem. Remediation strategies focus on using the wastes as raw materials to reduce storage costs and minimize the need for mining. An approach by vitrification reduces the volume of waste and inertizes hazardous elements by binding them to the structure of chemically stable glasses and glass-ceramics. The valorization process of sewage sludge by vitrification has been simulated by producing a glass and a glass-ceramic from a basalt enriched in calcium that lies between the stability fields of pyroxene and melilite in the system CaO-MgO-SiO

-Al

O

. Nucleation at the temperature of maximum nucleation rates (650 and 675 ºC) of this glass causes the formation of a biphasic system (crystal + glass) that constrains its rheological behavior enhancing the formation of a large amount of nuclei that result in a fine microstructure, forming a glass-ceramic. The microhardness of the glass (8.2 GPa) and the glass-ceramic (8.6 GPa) and leaching tests (in the ppb range) place both the glass and the glass-ceramics at the high end of the mechanical properties and chemical resistance of ceramic tiles for the building industry.

Keywords: glass-ceramic, sewage sludge, calcium, valorization, viscosity, nucleation

*Manuscript

Click here to view linked References

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

1. Introduction

Decontamination processes of wastewater daily generate huge volumes of sewage sludge, the solid fraction separated the treatment (including domestic septage) [1]. In Catalonia (population 7.5 M), water processing produced 119,230 tons of dry sewage sludge in 2015 [2]. The production of sludge peaked on the 2004–2008 period, averaging 140,000 tons per year [2] and decreased in a context of economic crisis that is currently receding. The disposal of these sewage sludge concerns two economically relevant issues: the availability of storage space in landfills and the environmental pollution that may derive from the sludge composition. An approach to solve this problem consists in designing new products using sewage sludge as a raw material – such as glasses and glass- ceramics for structural uses in the construction industry.

The preferred disposal of sewage sludge is its application in agricultural lands. However, this solution is limited to sludge that comply with the increasingly restrictive regulations established by the environmental governing bodies. Sewage sludge may contain concentrations of potentially toxic elements (PTE) that exceed by at least an order of magnitude those established by the European Economic Community for agricultural application [3], and Table 1. Hence, the design of new, environmentally safe products is crucial to optimize waste disposal processes. Vitrification is a valorization alternative where sewage sludge may be used as a raw material to produce glass appropriate for the building industry in the form of wall tiles or pavements [4]. Its advantages lie on both reducing the volume of the waste and binding of the components of the sludge – including the PTE – in the glass structure. Further inertization can be achieved using a glass-ceramic process, where the PTE will be emplaced in newly-formed mineral phases [5,6]. An inertization matrix analogous to sewage sludge may, once proven effective, be tested to bind industrial wastes such as galvanic or chromium sludge, which contain large concentrations of Zn and Ni, and Cr respectively.

Any potential applications are constrained by the bulk composition of sewage sludge, which largely

depends on its source area. Factors such as urban/rural origin, the actual processing undertaken at

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

microbial activity [7,8], long-term changes of soil uses of the watershed area, or the degree of groundwater contamination [9] influence the variability and the final composition of the sludge. The average composition of the inorganic fraction of urban sewage sludge (USS) may be simulated using basalt [4,10–16] and Table 1), the main differences being the concentration of Ca, Mg and P.

The addition of P to the basalt develops an increasingly sharp phase separation in the glass that results in partial crystallization, increasing the microhardness of the final products [17]. The addition of Ca is expected to improve the properties of the basalt glass by increasing microhardness and reducing viscosity.

Table 1. Composition of the basalt from Sant Joan les Fonts (measured using X-Ray fluorescence) compared to analyses of sewage sludge from WWTP reported on the literature [4,10–16]. The asterisks correspond to the values which exceed the range of real sludge compositions. The maximum concentrations are reported for the trace elements.

Oxide Basalt USS compositions Major components (wt%)

SiO

43.98 33 - 53.14

CaO 10.11 7.7 - 37.67

Al

O

14.16 11.73 - 21.48

FeO 10.88 4.64 - 10

MgO 10.05* 2.4 - 6.52

Na

O 3.31 0.6 - 5.2

TiO

2.51 1.19 - 2.9

K

O 1.98 0.9 - 3.8

P

O

0.55* 2.62 - 5.2

MnO 0.17 0.12 - 0.44

Trace elements (ppm)

Ba 585.4 330

Cr 263.5 1268

Cu 63.6 12701

Ni 172.9 1025

Pb 4.9 3519

Zn 86.2 31166

The properties of a successfully vitrified waste may be improved by a glass-ceramic process. A

glass-ceramic is any material produced from (partially) crystallized glass [18,19]. The strengths of

glass-ceramics as performing materials arise from their unique microstructure, achieved from the

understanding of the nucleation and crystallization processes. Controlling the composition of the

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

parent glass, the nucleation and crystallization processes and their effect on the rheological behavior is also essential to achieve low viscosity at high temperature, and to optimize the resistance to abrasion and chemical alteration [5,6]. Other interesting properties of glass-ceramics regarding their stability are the lack of porosity and high strength and toughness [18]. Ca-rich wastes are good candidates for vitrification and subsequent glass-ceramic processing - in products such as Sital slag from the steel and copper industries of Russia [20], cupola slag blast [21], blast furnace slag [22], tin tailings [23], molybdenum tailings [24], and tungsten tailings [25,26].

Xu and coworkers [27] studied the evolution of the nucleation rate on Na

O-2CaO-3SiO

glass and produced a glass-ceramic with fine-grained (crystal diameter about 10 µm), homogenous glass.

Their approach determined the temperatures of maximum nucleation rate using thermal analysis to establish the appropriate production procedure. The present study applies this method to the measurement of the temperatures of maximum nucleation rate of multiple phases on a glass representing a simplified composition of sewage sludge to produce a glass-ceramic with enhanced mechanical properties and chemical durability.

2. Material and methods 2.1. Glass synthesis

The production process of the parent glass (original composition) consists of mixing 84 wt% basalt from Boscarró old quarry (Sant Joan les Fonts, SJLFB, Catalunya) (Table 1) with 23 wt% reagent grade CaCO

(Panreac 121212) to provide the additional Ca. This composition should lie in the suitable workability range due to Ca´s expected decrease effect on the melting point of basalt [12].

The mixture is homogenized by crushing in a ball mill and then melted in a Pt crucible in a globular

alumina furnace equipped with SuperKanthal® heating elements. The melting process consisted of

heating the batch at 2 ºC/min up to 1000 ºC, followed by an isothermal step during 30 minutes to

complete decarbonation and further heating at 1 ºC/min up to 1450 ºC followed by an isothermal

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

is repeated twice to ensure homogenization. The whole batch made about 200 g of the parent glass.

Part of the glass is then annealed in a TechnoPiro® muffle furnace preheated at 500 ºC for 24 h to be able to cut a prism for the dilatometry measurements. No significant weight loss occurred during the production process.

2.2. Glass-ceramic production

The thermal treatment of the parent glass to obtain the glass-ceramic included the following steps based on the temperatures of maximum nucleation rate (T

) determined during the study:

- Heating up to 650 ºC at 2 ºC/min for 6 h (T

of magnetite)

- Heating at 675 ºC at 2 ºC/min for 6 h (2

T

, discussed further in the manuscript) - Heating up to 874 ºC at 10 ºC/min for 6 (the temperature of the exothermic DTA peak) - Heating to 1000 ºC at 2 ºC/min for 6 h (at this temperature merwinite has been exhausted by

reacting with diopside to form akermanite) - Free cooling

2.3. Chemical composition

The chemical composition of the basalt has been measured by X-Ray Fluorescence using a sequential X-ray spectrophotometer Phillips PW2400. The major elements have been measured in lithium tetraborate pearls and the trace elements in pressed pellets.

The chemical composition of the parent glass has been measured in a JEOL JXA 8230 electron

microprobe (EMPA). Quantitative electron microprobe analyses have been obtained in wavelength-

dispersive spectroscopy (WDS) mode, operating with an accelerating voltage of 20 kV, a beam

current of 16 nA for Al, Si, Ti, Ca, K, Mg, Fe, Mn, 8nA for Na and 6 nA for P, and a beam diameter

of 5 μm. Corundum (AlKα), wollastonite (SiKα, CaKα), rutile (TiKα), orthoclase (KKα), apatite

(PKα), periclase (MgKα), albite (NaKα), synthetic Fe2O3 (FeKα) and rhodonite (MnKα) have been

used as standards. The employed analyzing crystals have been TAP for Al and Si, PETJ for Ti, Ca,

K and P, TAPH for Mg and Na and LIFH for Fe and Mn.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

2.4. Raman spectroscopy

The Raman spectrum of the parent glass has been obtained with a T64000 Jobin-Yvon Raman spectrometer equipped with a CCD detector. The light source was an Ar+ ion laser operating at 488 nm with a typical output of 100 mW on the sample. The integration time was 900 s and the spectral range was between 100 – 1500 cm-

. The spectrum has been treated with the Long correction (Long, 1977, Neuville and Mysen, 1996) and normalized to the total area [29,30].

2.5. Thermal analysis

Glass thermal evolution has been studied by Differential Thermal Analysis (DTA) in a Netzsch DTA-TG STA 409C equipment. A preliminary analysis is made on glass powder (particle size <50 µm) at 10 ºC/min in an alumina crucible under a 70 mL/min N

flow and using pure Al

O

(Perkin- Elmer 0419-0197) as a reference material to determine the crystallization events. The small particle size ensures full crystallization of the sample. The rest of the as-quenched glass is then ground and screened to a particle size between 400 and 500 µm to ensure bulk crystallization. Further DTA measurements are performed using about 70 mg of glass sample.

The glass particles are heated at 15 ºC/min up to nucleation temperatures (T

) between 500 – 850 ºC during 3 h and then further heated up to 1350 ºC before cooling. These treatments to different T

cause a shift of the temperature of the exothermic DTA peak attributed to crystallization. The temperature of the maximum nucleation rate is then determined from the plot of the inverse of the temperature of the exothermic peak as a function of the nucleation temperature [27].

2.6. X-Ray diffraction

The presence of amorphous and crystalline phases in the glass and after thermal treatments is

assessed by X-Ray Diffraction (XRD) obtained in a PANalytical X’Pert PRO MPD Alpha1 powder

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

1.5406 Å), work power 45 kV – 40 mA, scanning range 4 – 80º 2 with step size of 0.017º and measuring time 50 s. In-situ X-Ray Diffraction (XRD) under vacuum atmosphere was carried out in an Anton Paar HTK1200N High Temperature chamber (HT-XRD) coupled to the previously described equipment. The experiment consisted in heating powdered glass at a rate of 20 °C/min with 1h isothermal steps at 28, 300, 500, then from 550 to 1200 ºC, and 28 ºC after cooling – 7 spectra were taken during each one of these steps. Data from the HT-XRD patterns were used to determine how the intensity of certain peaks varies with temperature as well as to calculate the coherent size of the crystallites of each mineral phase nucleated during the thermal treatment from Scherrer’s equation (Eq. 1) [31]:

[Eq 1]

In this equation τ stands for the average size of the crystalline domains; K is a dimensionless shape (0.9), λ is the X-Ray wavelength (1.5406); β is the full width of the peak at half maximum minus the instrumental line broadening, in radians; and θ is the Bragg angle – in radians.

The quantification of the mineral phases of the glass-ceramic was obtained from the Rietveld refinement of the XRD non-oriented powder diffraction on FullProf v 3.00 software [32].

2.7. Dilatometry

The dilatometric curve of the glass is obtained in a Linseis L76/1550 horizontal dilatometer at a heating rate of 10 ºC/min and air flow of 5 mL/min. The glass transition temperature (T

) is calculated from the curve using the tangents method. The coefficient of thermal expansion is calculated between 100 and 500 ºC.

2.8. Viscosity measurements

The effect of nucleation on the viscosity of glass is investigated by hot-stage microscopy (HSM)

prototype [33]. The studied samples are nucleated at 650 ºC in the DTA furnace according to the

following sequence: the glass is heated at 10 ºC/min up to 500 ºC, where the temperature is

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

stabilized for 10 min and later increased at 20 ºC/min up to the T

for a period of time ranging from 0.5 to 8 h. The temperatures of the fixed viscosity points are then determined in glass powder (diameter < 45 µm) cylinders produced in a uniaxial press using a 1/20 solution of Elvacite® in acetone. The deformation of the probe during heating at 5 ºC/min between 25 and 1450 ºC is recorded using the ProgRes CapturePro 2.8.8. software. The images corresponding to the fixed viscosity points have been identified using Hot-Stage software [33]. The viscosity-temperature curves have been plotted using the temperatures of the fixed viscosity points determined during the HSM analysis and the viscosity values from Pascual and coworkers [34].

2.9. Electron microscopy

Textural information and qualitative punctual chemical analysis are obtained by scanning electron microscopy (SEM) and Energy-Dispersive X-Ray Spectroscopy (EDX) using a JEOL J-7100 field emission scanning electron microscope with EDS detector and backscattered electron detector (BDS).

2.10. Microhardness Vickers

The Vickers hardness (VH) of the samples is measured on polished glass probes using a Galileo Isoscan OD Vickers microindenter with a load of 294 N.

2.11. Leaching tests

Elementary analysis of the leachates serves to evaluate the stability of the glasses according to DIN

38414-S4 [35]. The test is performed in 10 g of sample – particle size between 2 and 4 mm – dried

at 50 ºC and mixed in 100 mL of deionized water. The mixture is agitated at room temperature

during 24 h and the liquid is separated from the solid using a 0.45 µm pore size filter. The leachates

were analyzed by inductively coupled plasma optical emission spectrometry (ICP-OES, Optima

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

3100×, PerkinElmer) and inductively coupled plasma mass spectrometry (ICP-MS, Elan 6000, PerkinElmer).

3. Results

3.1. Original glass characterization

The chemical composition of the parent glass was elaborated from a basalt doped with calcium (Table 2). This mixture is comparable to USS compositions gathered from the literature – as explained in the introduction – in spite of some minor differences. It is enriched in Mg, and specially impoverished in P and Ca when compared to sludge [17,36]. The parent glass can be described as a calcic silicate with high Fe contents, which cause its dark brown color in thin section, black in bulk.

Table 2. Chemical composition of the parent glass B16Ca obtained by EMPA. The values in brackets correspond to the standard deviation of the measurements.

The Raman spectrum of the parent glass may be divided in 3 parts (Figure 1). The low frequency range includes the boson peak between 50 and 150 cm

, linked to the rotation of silicate tetrahedra [37]. The high intensity region of the spectrum contains three broad bands: the low frequency envelope (LF), between 250 and 630 cm

, the middle frequency envelope (MF) between 630 and 785 cm

and the high frequency envelope (HF) between 735 and 1250 cm

(Figure 1). The LF envelope is related to vibrations of bridging oxygen in rings of tetrahedra (McMillan and Piriou, 1982; Mysen et al., 1980; Neuville, 2006; Neuville et al., 2014; Neuville and Mysen, 1996;

wt% SiO

Al

O

Fe

O

CaO MgO Na

O TiO

K

O P

O

MnO

B16Ca

36.11 (0.34)

12.19 (0.17)

10.06 (0.29)

24.44 (0.33)

9.19 (0.56)

2.28 (0.10)

2.02 (0.08)

1.12 (0.07)

0.46 (0.10)

0.15

(0.03)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

Pasquarello et al., 1998; Seifert et al., 1982; Umari et al., 2003). The MF corresponds to the intertetrahedral bending mode of polymerized species [45,46]. The HF envelope is attributed to the vibrations of T-O- bonds – where T stands for the fourfold coordinated cations Si

, Al

, Fe

or P

, and O

for the non-bridging oxygen (NBO) – and the structural effect of network-modifying and charge-balancing cations [see references in [39], and [30,41,46–50]. All the vibrations of the HF region are convoluted in a single wide envelope. The ratio between the area of the HF and the LF envelopes shows that the glass is essentially depolymerized.

Figure 1. Unpolarized normalized Raman spectrum of the parent glass.

The thermal behavior of the original glass shows two exothermal events attributed to the formation

of different mineral phases (Figure 2). The first signal is a sharp peak at 874 ºC and the second is

smaller, wider and located at 924 ºC. An endothermic event at 1128 ºC is linked to the melting of

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

the system (melting of akermanite). Glass transition temperature measured at 10 ºC/min of the parent glass is 651 ºC and its thermal expansion coefficient is 9.7·10

ºC

.

Figure 2. DTA plot of the original glass obtained using a heating rate of 10 ºC/min.

3.2. Determination of the temperatures of maximum nucleation rate and crystal growth

The temperature of maximum nucleation rate can be obtained from the shift of the exothermic peak

corresponding to crystallization when the glass is treated at different nucleation temperatures (T

)

[27]. This approach should supply T

for each of the mineral phases that crystallize from the

original glass. Two different events have been isolated. The first is a broad peak centered at 650 ºC

from the shift of the exothermic peak at 874 ºC (Figure 3), which appears in all the analyses. The

second is a narrower peak centered at 675 ºC, which is only detected for T

below 750 ºC,

corresponding to the variation of the peak at 924 ºC (Figure 3).

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

Figure 3. Plots representing the shift of the inverse of the temperature of the exothermic peak as a function of temperature corresponding to the two exothermic events.

The sequence of nucleation of the newly-formed minerals and the attribution of the T

are studied from the evolution of the intensity of the main diffraction peak after a sudden heating at T

and further heating up to the temperature of the exothermic event (T

) from DTA to induce crystal

growth (Table 3 and Figure 4). The first T

is attributed to the nucleation of magnetite (Fe

O

), as

it has the most intense XRD peaks at 650 ºC; its peak height later diminishes and becomes

undetectable at 700 ºC. Nepheline (NaAlSiO

) starts nucleating at 650 ºC and its nucleation rate is

larger at 700 ºC, but cannot be proved to be at its maximum at this temperature (Figure 4). The

nucleation of diopside (CaMgSi

O

) and merwinite (Ca

MgSi

O

) is fast and at about the same

temperature range: between 650 ºC and 700 ºC. The proximity between all these nucleation events

makes it difficult to attribute the T

linked to the second event at 675 ºC to a single phase with

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

Table 3. Intensities of the main diffraction peaks of the newly-formed mineral phases after sudden thermal treatments at the T

followed by growth at the temperature of the first exothermic. The numbers below the mineral phases correspond to the Powder Diffraction File (PDF) used in the Rietveld refinement.

The intensities and narrowness of the XRD peaks at 750 and 800 ºC, coupled with the apparent scarcity of residual glass, show that the nucleation stage has finished and crystal growth is the current /ongoing process. A new mineral phase, akermanite (Ca

Mg[Si

O

],) is detected at 750 ºC.

Figure 4. Sequence of XRD patterns obtained after a sudden heating of glass samples at a) T

and b) T

for each phase. The temperatures in the plot correspond to the T

of each treatment.

3.3. Evolution of the crystalline phases in the devitrification process

XRD spectra show the evolution of the newly-formed mineral phases between 770–1200 ºC during devitrification of the original glass in in-situ HT-XRD. The sample stays amorphous up to 760 ºC and the first diffraction peaks can be detected at 770 ºC. The nucleation process is not controlled in

T

(ºC)

Newly-formed minerals (cps) Magnetite

01-088-0315

Nepheline 01-076-2467

Diopside 01-072-1379

Merwinite 01-089-2432

Akermanite 01-079-2424

600 113 - - - -

650 301 63 431 - -

700 165 167 265 - -

750 - 1726 6001 2697 33

800 - 1450 5892 2218 1893

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

this experiment, hence the series of XRD patterns show both nucleation and growth of the minerals with increasing temperature (Figure 5).

Figure 5. HT-XRD spectra in the range 2 between 20 and 40 º.

Diopside, nepheline and merwinite can be seen in the HT-XRD spectrum at 770 ºC, whereas

akermanite is detected from 800 ºC (Figure 5). The evolution of the intensity of each of the mineral

phases has been tracked using the following diffraction peaks: d

for nepheline, d

for merwinite,

d

for diopside and d

for akermanite (Figure 6). They are the most intense that did not overlap

with peaks from coexistent phases. The evolution of magnetite could not be tracked in this

experiment because its main diffraction peak overlaps with a major peak of diopside and the rest of

reflections are not intense enough.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

Figure 6. Evolution of the intensities of selected XRD peaks (d

for nepheline, d

for merwinite, d

for diopside and d

for akermanite) as a function of temperature.

Nepheline is stable between 760 ºC and 1080 ºC, with a maximum intensity at 860 ºC. Merwinite has a similar behavior: it also reaches a maximum at 860 ºC and becomes undetectable over 980 ºC.

The peak intensities of akermanite increases during the decrease of these two phases, reaches its maximum at 970 ºC and becomes undetectable at 1190 ºC. The peak intensity of diopside increases along the whole temperature range, alongside with the crystallization and melting reactions of merwinite and akermanite (Figure 5 and Figure 6). The sample does not undergo additional changes during cooling.

The sizes of coherent crystals at each temperature can be calculated using Scherrer’s equation to

establish the growth of each phase. The maximum crystal size reached in HT-XRD for all the

minerals is between 40 – 70 nm. This can be due to the fact that all phases nucleate in a narrow

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

temperature range, hence forming a large number of nuclei before reaching the crystal growth temperature (T

) and constraining crystal growth. Magnetite crystals are still smaller, reaching a maximum of 18 nm. Its role is to act as a nucleus for the growth of diopside and it is absorbed inside its structure to approach the Fe-rich end-member.

3.4. Evolution of the viscosity of the original glass as a function of nucleation time

The viscosity-temperature curves of samples treated at a nucleation temperature of 650 ºC during different periods of time (between 0 and 8 h) are traced using the fixed viscosity points from HSM [34] (Figure 7). In general, a longer nucleation time increases the viscosity (for a nucleation period between 0 and 4h), in the high viscosity range (between 10

Pa·s). However, the sample treated during 8 h is very similar to the sample nucleated for 2 h as the system becomes multiphasic with the separation between two solid phases, one crystalline and one amorphous. However, increasing nucleation time reduces the viscosity in the low viscosity range (between 10

Pa·s). At 650 ºC the phases that will nucleate are magnetite and nepheline. The nucleation of nepheline will extract Na and K from the glass, thus increasing the viscosity according to longer nucleation times. The variation of viscosity is also conditioned by the effect of nucleation on the sintering process.

Between the first shrinkage (10

Pa·s) and the softening points (10

Pa·s), all matter will be in the solid state, either as a crystalline or an amorphous phase. The shrinkage of the probe at this stage is controlled by an evaporation-condensation process depending on vapor pressure. Increasing the amount of nuclei will reduce vapor pressure because crystals have lower vapor pressure than amorphous materials. The formation of nuclei requires increasing the vapor pressure by means of a temperature rise for the sintering process to occur, which is in good agreement with the temperature at which each sample reaches the maximum shrinkage point (10

Pa·s) increasing with nucleation time.

For viscosities below the softening point, matter will be a mixture of solid crystals and a melt. In

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

between solid and liquid [51]. In addition, a larger number of nepheline crystals will cause a more sudden decrease in viscosity upon its melting between 897 and 1080 ºC. In the viscosity- temperature curves this corresponds to a drop between first shrinkage and softening, which can be correlated to the decrease in the viscosity of the melt caused by the incorporation of Na and K that have left the nepheline structure. The second, more gradual, viscosity drop is attached to a decrease in the crystal/liquid ratio, which is independent of nucleation time, as the curves are approximately parallel in this range (Figure 7).

Figure 7. Viscosity-temperature curves for glasses with a) different nucleation times b)

magnification focusing on the samples nucleated before the HSM experiments. The highlighted

areas represent the biphasic region, where a crystal and a glass/melted phase coexist (purple), the

nucleation range, which are XRD-amorphous (red) and the growth range, where crystallization

increases viscosity as a function of the nucleation time (green).

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

3.5. Glass-ceramic characterization

The glass-ceramic has been obtained through the thermal treatment of the parent glass based on the T

and T

exposed earlier. It is formed by nepheline (12.15 wt%), magnetite (6.03 wt%), diopside (41.32 wt%) and akermanite (40.50 wt%) (Figure 8).

Figure 8. XRD pattern of the glass-ceramic showing the observed intensity (Yobs) of the

experiment, the calculated intensity of the Rietveld refinement (Ycalc), the difference curve and the Bragg positions of the diffraction for each mineral phase. The labels identify the 100 % intensity diffraction peak of each crystalline phase.

Treating the parent glass at the T

generates a large number of nuclei, which grow upon further

heating. Between each T

the heating is fast to minimize the growth of the phases that were

nucleated in previous steps, ensuring a homogeneous crystal size. The bulk microstructure of the

glass-ceramic observed on SEM consists on idiomorphic crystals ranging between 60 and 120 nm

(Figure 9). Diopside, akermanite and nepheline form a homogeneous grain microstructure.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

than in akermanite). The observed sizes of the crystals are larger than 40–70 nm obtained from Scherrer’s equation for diopside, nepheline and akermanite and 18 nm for magnetite. This difference arises from the steps for crystallization during glass-ceramic production being much longer than for in-situ HT-XRD.

Figure 9. SEM micrographs showing the texture of the glass-ceramic. a) a BDS image of the mass of the crystals where dense magnetite crystals stand out b) a detail of the mixture of fine grained idiomorphic diopside, akermanite and nepheline.

3.6. Microhardness

The Vickers microindentation test has given a microhardness of 8.2 GPa for the parent glass and of

8.6 GPa for the glass-ceramic. These values lie in the hardness range of the diopside-hedenbergite

series (from 7.7 to 9.8 GPa), are higher than the augite series (from 6.6 to 8.0), and considerably

higher than the hardness of akermanite (5.2 to 6.6 GPa). The abundance of diopside is the main

contributor to the increase of the microhardness, which leads to an improvement on the mechanical

properties of the obtained material.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

3.7. Leaching

The glass-ceramic process has increased the overall stability of the glass, as the concentration of most of the elements in the glass-ceramic leachate are lower than in the glass leachate (Table 4).

However, the leachability of elements such as Ca, Si, Zn and Pb, As and Cd is higher from the glass-ceramic than the glass. The concentration of each element in the leachates is several orders of magnitude below their concentrations in the original basalts.

Table 4. Concentration of selected elements on the leachates after blank subtraction. BB stands for

“below blank concentration” and BDL for “below detection limit”.

[ppm] in solution Ca K Si Na Mg Fe Al Ba P

Glass 3.52 2.24 4.17 9.73 2.73 0.39 1.18 0.04 BB

Glass-ceramic 4.43 1.88 4.79 8.31 2.45 0.11 0.88 0.09 BB

[ppb] in solution Zn Pb As Cu Cr Ni Cd Ti

Glass BB 0.83 0.23 11.09 BDL 9.00 BB 76.32

Glass-ceramic 2.85 1.04 0.71 10.68 BDL 7.16 0.18 26.73

DIN 38414-S4 [30] 4000 500 500 2000 500 400 40 N/A

4. Discussion

The composition of the parent glass corresponds to the border between the stability fields of

pyroxene and melilite in the section at 10 wt% Al

O

of the phase diagram of the system CaO-

Al

O

-MgO-SiO

, the major glass components. The crystallization of akermanite – the magnesian

end-member of the melilite solid solution ((Ca,Na)(Al,Mg,Fe

)[(Al,Si)SiO

]) – and its relationship

with merwinite is a controversial question in the literature [22] consider merwinite a metastable

phase formed during the thermal treatments of multicomponent glasses in the initial stages of the

devitrification process, around T

. The relaxation of the glass structure upon at higher temperatures

will facilitate the diffusion of ions [14] causing the destabilization of merwinite, which will react

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

A study of the crystallization of a glass of a composition in the akermanite-gehlenite system (2CaO·(1-x)·MgO·xAl

O

·(2-x)SiO

) proved that merwinite is a metastable phase for all x below or equal to 0.6 and attributed this fact to structural reasons [53]. In the case of the melilite minerals, corresponding to the sorosilicate group, the structure consists in a fragmented network of silica tetrahedra and modifier cations are stabilized by the ionic bonds between the tetrahedra and the cations [22]. Merwinite is a nesosilicate and thus presents a lower concentration of silica tetrahedra, creating a less strong network that melts at lower temperature [53]. The formation of akermanite may then be explained from the diffusion of calcium from merwinite to diopside described by the following reaction:

Merwinite + Diopside  Akermanite

3CaO·MgO·2SiO

+ CaO·(MgO,FeO)·2SiO

The bulk composition of the parent glass can be expressed as 2CaO·1.05MgO·0.55Al

O

·2.76SiO

using the formula of melilite. In this case, x is effectively below 0.6 and then merwinite is metastable. Moreover, the already insufficient concentration of Al in the system is also influenced by the formation of nepheline during the devitrification process. Nepheline is the phase with a higher Al concentration between those formed during the devitrification process hence its formation extracts Al from the system, destabilizing merwinite and enhancing the crystallization of akermanite. Even the melting of nepheline over 860 ºC does not introduce enough Al in the system to allow the formation of merwinite, and thus it is absent from the glass-ceramic.

Nucleation and crystallization influence macroscopic properties of the melt such as viscosity [54],

which are essential in the production of the glass and the glass-ceramic. The large increase in

viscosity between the parent and the nucleated glasses is attributed to the formation of a biphasic

system consisting in a crystalline and an amorphous phase. The crystals facilitate the formation of

the fine-grained glass-ceramic by hampering the viscous flow at the nucleation range, limiting the

reactivity between the different phases (Figure 7) in a similar manner. A further temperature rise

decreases the bulk viscosity of the material permitting reactions between minerals but preserving

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

the fine microstructure until the melting range (around 10

Pa·s), where the system becomes monophasic again. Villeneuve and coworkers observed a similar behavior in their viscosity measurements during nucleation processes on basaltic melts of Piton de la Fournaise volcano [55].

The production process of the glass-ceramic based on the determination of the temperatures of maximum nucleation rates and the study of crystal growth has provided a crystalline material with a fine microstructure. Obtaining extensive nucleation on a glass-ceramic based on basalt usually requires a Fe

O

/FeO ratio over 0.5 [56]. The raw basalt Fe

O

/FeO equals 0.501 [57] and supplementary oxidation is provided by melting the ground basalt under air atmosphere [18]. This fine microstructure is a direct result of the treatment of the glass at the temperatures of maximum nucleation rate for an extended period (in this case, 6 h) followed by a fast heating of the sample up to the growth temperature to avoid the formation of metastable phases such as merwinite. An additional factor that may help in developing this microstructure is the presence of pre-structured domains as hinted by the noisy Raman signal. The presence of magnetite and diopside in the mineralogy of the glass-ceramic shows a good potential for the inertization of PTE such as Cr or Zn present in sludge [10–12] because they could be hosted in the spinel structure or in the pyroxene lattice.

The microhardness of the glass and the glass-ceramic lie on the same range, and near the higher

limit, of the reported values for basalt glasses and glass-ceramics in the literature (Table 5). The

extensive nucleation of the parent glass creates a fine microstructure and limits the eventual

formation of porosity due to crystal growth, making the glass-ceramic harder than the glass in spite

of bearing approximately 40 wt% of the soft mineral akermanite (5.2–6.6 GPa). These results have

to be taken with certain reserve because the difference in hardness between B16Ca glass and glass-

ceramic is within the experimental error. However, the trend of increasing hardness is a promising

advocate for the glass-ceramic process as a means for improving the mechanical properties of

multicomponent glass.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

Table 5. Microhardness results of this study compared to Vickers microhardness values for basaltic glasses and glass-ceramics in a [58], b [59], and c [60].

Material Vickers Microhardness (GPa) Sant Joan les Fonts Basalt glass 7.7  0.3

B16Ca glass 8.2  0.5

B16Ca glass-ceramic 8.6  0.5

Canary Island Basalt glass (a) 5.16 – 6.28 Canary Island Basalt glass-ceramic (a) 6.62 – 8.66

Basalt rock fiber glass (b) 7.7

Basalt glass/crystal mixtures (b) 6.8 – 8.9

Holyoke basalt glass (c) 8.9

Glass-ceramics are generally characterized by a good chemical resistance, comparable to other ceramic materials [61]. In this study, the leachability tests provided complex results. The decrease in the concentrations of most of the main components between the leachates of the glass and the glass-ceramic point to an overall better inertization potential. However, the leachability of Ca, Si, Zn, Pb, As and Cd is lower in the glass than in the glass-ceramic. Fredericci and coworkers observed a similar situation in the increase of weight loss between a blast furnace slag glass and glass-ceramics [22]. This situation might be caused by the high amount of akermanite, which is the phase with higher Ca and Si contents of the glass-ceramic. Karamanov and coworkers produced a diopside glass-ceramic which underwent a weight loss as low as 0.3 wt% [62], whereas 1.2 wt%

losses have been reported for acid-resistant melilite glass-ceramics [22]. Therefore, a line emerging from this research could focus on decreasing the amount of Ca to increase the proportion of pyroxene and maximize the chemical resistance of the glass-ceramic.

5. Conclusions

We achieved the vitrification of a Ca-rich basalt analogous to sewage sludge and its transformation

into a diopside-akermanite-magnetite-bearing glass-ceramic based on the determination of the

temperatures of maximum nucleation rate and crystal growth of the newly-formed phases. The

glass-ceramic process enhances the microhardness of the product, which lies in the upper part of the

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

literature values for ceramic tiles where abrasion-resistant materials are required such as the building industry. Optimizing the leachability in the glass-ceramic may require adjusting the concentration of calcium to reduce the crystallization of akermanite, promoting the formation of more spinel-like phases and pyroxene. Concerning the production process, the apparent viscosity of the system increases as it becomes biphasic during nucleation and crystallization. This confers a certain rigidity to the material, limiting the reactions between formed nuclei and the residual liquid phase until crystallization is finished. The result is a nanometric homogeneous microstructure responsible for the enhanced mechanical properties of the glass-ceramic. Both the glass and the glass-ceramic effectively bind the PTE in their structures in compliance with European legislation – in the glass structure and the crystalline lattices of the spinel-group and pyroxene minerals respectively. The designed process opens the way to use sewage sludge-like waste as raw materials for environmentally safe products, with the additional advantage of reducing the need for their mining.

6. Acknowledgements

This research was supported by Consolidated Group for Structure and Materials Desing , 2017SGR1687 and by the Fundació Bosch i Gimpera Project 307466. The authors would like to thank the staff of the Centres Científics i Tecnològics of the University of Barcelona (CCiTUB) for their technical support and Esther Vilalta and the Departament de Ciència de Materials i Enginyeria Metal·lúrgica for the access to the Vickers microindenter. M. Tarragó received support from a PhD grant from the Ministerio de Educación, Cultura y Deporte (FPU13/04507).

7. References

[1] EPA, Land Application of Sewage Sludge, Fed. Regist. (1994).

http://water.epa.gov.akadns.net/polwaste/wastewater/treatment/biosolids/upload/2002_06_28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

ds/upload/2002_06_28_mtb_biosolids_sludge.pdf.

[2] Agència Catalana de l’Aigua, Evolució de les dades de la producció i de la gestió dels fangs ., (2015). https://aca-

web.gencat.cat/aca/documents/ca/actuacions/sistemes_sanejament/com_funcionen_edars/evo l_fangs_web.pdf (accessed February 6, 2017).

[3] V.J. Inglezakis, A.A. Zorpas, A. Karagiannidis, P. Samaras, I. Voukkali, S. Sklari, EUROPEAN UNION LEGISLATION ON SEWAGE SLUDGE MANAGEMENT, 23 (2016) 635–639.

[4] M. Romero, J.M. Rincón, Microstructural characterization of a goethite waste from zinc hydrometallurgical process, Mater. Lett. 31 (1997) 67–73. doi:10.1016/S0167-

577X(96)00235-2.

[5] M. Garcia-Valles, G. Avila, S. Martinez, R. Terradas, J.M. Nogues, Heavy metal-rich wastes sequester in mineral phases through a glass-ceramic process, Chemosphere. 68 (2007) 1946–

1953. doi:10.1016/j.chemosphere.2007.02.034.

[6] M. Tarrago, M. Garcia-Valles, M.H.H. Aly, S. Martinez, Valorization of sludge from a wastewater treatment plant by glass-ceramic production, Ceram. Int. 43 (2017) 930–937.

doi:10.1016/j.ceramint.2016.10.083.

[7] T.A. Guillemet, P. Maesen, E. Delcarte, G.C. Lognay, A. Gillet, J.J. Claustriaux, M. Culot, Factors influencing microbiological and chemical composition of South-Belgian raw sludge, Biotechnol. Agron. Soc. Environ. 13 (2009) 249–255.

http://www.scopus.com/inward/record.url?eid=2-s2.0-

70049084577&partnerID=40&md5=6d80b3f51734780da3affb02c32b2494.

[8] M.I. Ansari, A. Malik, Seasonal variation of different microorganisms with nickel and cadmium in the industrial wastewater and agricultural soils, Environ. Monit. Assess. 167 (2010) 151–163. doi:10.1007/s10661-009-1038-y.

[9] N. Hecht, D. Duvall, Characterization and Utilization of municipal and utility sludges and

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

ashes, 1975.

[10] M.K. Hossain, V. Strezov, P.F. Nelson, Thermal characterisation of the products of wastewater sludge pyrolysis, J. Anal. Appl. Pyrolysis. 85 (2009) 442–446.

doi:10.1016/j.jaap.2008.09.010.

[11] T. Tervahauta, S. Rani, L. Hernández Leal, C.J.N. Buisman, G. Zeeman, Black water sludge reuse in agriculture: Are heavy metals a problem?, J. Hazard. Mater. 274 (2014) 229–236.

doi:10.1016/j.jhazmat.2014.04.018.

[12] R.D. Rawlings, J.P. Wu, A.R. Boccaccini, Glass-ceramics: Their production from wastes-A Review, J. Mater. Sci. 41 (2006) 733–761. doi:10.1007/s10853-006-6554-3.

[13] S. Suzuki, M. Tanaka, T. Kaneko, Glass-ceramic from sewage sludge ash, J. Mater. Sci. 2 (1997) 1775–1779. doi:10.1023/a:1018584202392.

[14] M. Erol, S. Küçükbayrak, A. Ersoy-Meriçboyu, Production of glass-ceramics obtained from industrial wastes by means of controlled nucleation and crystallization, Chem. Eng. J. 132 (2007) 335–343. doi:10.1016/j.cej.2007.01.029.

[15] S. Do Yoon, Y.H. Yun, Preparation of glass ceramics from sludge bottom ash and waste glass, J. Ceram. Process. Res. 12 (2011) 361–364.

[16] J.M. Juoi, D. Arudra, Z.M. Rosli, K. Hussain, A. Japper Jaafar, Microstructural properties of glass composite material made from incinerated scheduled waste slag and soda lime silicate (SLS) waste glass, J. Non. Cryst. Solids. 367 (2013) 8–13.

doi:10.1016/j.jnoncrysol.2013.02.004.

[17] M. Tarrago, M. Garcia-Valles, S. Martínez, D.R. Neuville, Phosphorus solubility in basaltic glass: Limitations for phosphorus immobilization in glass and glass-ceramics, J. Environ.

Manage. 220 (2018). doi:10.1016/j.jenvman.2018.04.079.

[18] W. Holand, G.H. Beall, Glass-ceramic Technology, 2012.

doi:10.1002/9781118265987.fmatter.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

and phase separation: from research to applications, 2017.

[20] N. Pavluskin, Vitrokeramik. Grundlagen der Technologie, 1986.

[21] G. Agarwal, R.F. Speyer, Devitrification hardening of cupola slag glass with CaO and SiO

additions, J. Non. Cryst. Solids. 135 (1991) 95–104. doi:10.1016/0022-3093(91)90409-Y.

[22] C. Fredericci, E.D. Zanotto, E.C. Ziemath, Crystallization mechanism and properties of a blast furnace slag glass, J. Non. Cryst. Solids. 273 (2000) 64–75. doi:10.1016/S0022- 3093(00)00145-9.

[23] J.R.H. Arancibia, P. Alfonso, M. García-Valles, S. Martínez, D. Parcerisa, C. Canet, F.M.

Romero, Obtención de vidrio a partir de residuos de la minería del estaño en Bolivia, Bol. La Soc. Esp. Ceram. y Vidr. 52 (2013) 143–150. doi:10.3989/cyv.192013.

[24] J. Shi, H. Feng, C. Ye, L. Hu, J. Xie, H. Yang, X. Liu, Preparation and characterization of CaO·Al

O

·SiO

glass-ceramics from molybdenum tailings, Mater. Chem. Phys. 197 (2017) 57–64. doi:http://dx.doi.org/10.1016/j.matchemphys.2017.05.028.

[25] P. Alfonso, D. Castro, M. Garcia-Valles, M. Tarrago, O. Tomasa, S. Martinez, Recycling of tailings from the Barruecopardo tungsten deposit for the production of glass, J. Therm. Anal.

Calorim. (2016). doi:10.1007/s10973-016-5332-y.

[26] P. Alfonso, O. Tomasa, M. Garcia-Valles, M. Tarrago, S. Martínez, H. Esteves, Potential of tungsten tailings as glass raw materials, Mater. Lett. 228 (2018).

doi:10.1016/j.matlet.2018.06.098.

[27] X.J. Xu, C.S. Ray, D.E. Day, Nucleation and crystallization of Na2O-2CaO-3SiO2 glass by differential thermal analysis, J. Am. Ceram. Soc. 74 (1991) 909–914. doi:10.1111/j.1151- 2916.1991.tb04321.x.

[28] D.A. Long, Raman spectroscopy, McGraw-Hill, 1977.

[29] C. Le Losq, D.R. Neuville, R. Moretti, J. Roux, Determination of water content in silicate glasses using Raman spectrometry: Implications for the study of explosive volcanism, Am.

Mineral. 97 (2012) 779–790. doi:10.2138/am.2012.3831.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

[30] C. Le Losq, D.R. Neuville, P. Florian, G.S. Henderson, D. Massiot, The role of Al3+ on rheology and structural changes in sodium silicate and aluminosilicate glasses and melts, Geochim. Cosmochim. Acta. 126 (2014) 495–517. doi:10.1016/j.gca.2013.11.010.

[31] A. Monshi, M.R. Foroughi, M.R. Monshi, Modified Scherrer Equation to Estimate More Accurately Nano-Crystallite Size Using XRD, World J. Nano Sci. Eng. 02 (2012) 154–160.

doi:10.4236/wjnse.2012.23020.

[32] J. Rodríguez-Carvajal, Recent advances in magnetic structure determination by neutron powder diffraction, Phys. B Condens. Matter. 192 (1993) 55–69. doi:10.1016/0921- 4526(93)90108-I.

[33] M. Garcia-Valles, H.S. Hafez, I. Cruz-Matias, E. Verges, M.H. Aly, J. Nogues, D. Ayala, S.

Martinez, Calculation of viscosity-temperature curves for glass obtained from four wastewater treatment plants in Egypt, J. Therm. Anal. Calorim. 111 (2013) 107–114.

doi:10.1007/s10973-012-2232-7.

[34] M.J. Pascual, L. Pascual, A. Duran, Determination of the viscosity-temperature curve for glasses on the basis of the fixed viscosity points determined by hot stage microscopy, Phys.

Chem. Glas. 42 (2001) 61–66.

[35] DIN-38414S4, Deutsche Einheitsverfahren zur Wasser, Abwasserund

Schlammuntersuchung, Bestimmung der Eluierbarkeit von Wasser (S4), in: 1984.

[36] M. Tarrago, H. Esteves, M. Garcia-Valles, S. Martínez, D.R. Neuville, Effect of Ca in P- doped basaltic glass-ceramics: Application to waste inertization, Mater. Lett. 220 (2018) 266–268. doi:10.1016/j.matlet.2018.03.020.

[37] B. Hehlen, E. Courtens, A. Yamanaka, K. Inoue, Nature of the Boson peak of silica glasses from hyper-Raman scattering, J. Non. Cryst. Solids. 307–310 (2002) 87–91.

doi:10.1016/S0022-3093(02)01444-8.

[38] P.F. McMillan, B. Piriou, The structures and vibrational spectra of crystals and glasses in the